5 Web Optimization Tips
As they say, the devil is in the details. Often times, there are too many details within testing and analytics. It's best to keep it simple, stupid.
However, in the online school of hard knocks over 15 years, I've learned some shareworthy tips for gathering data online and offline when performing online tests.
Use Archive.org
Archive.org, a.k.a. the "Wayback Machine" has been around forever in Web terms.
It's old... It's probably the oldest, but one of the best visual documentation source on the Web. I've used it several times a year for the past, I don't know 15 years. I used it successfully in 2016 and then already in 2017.
In fact, someone told me a story this week that Google contacted them back in 2000 to imply that said company infringed on Google's new simplistic home page layout. This person then referred Google to archive.org to explain that the opposite was true. Google offered their mea culpa and then added this company to the list of early adopters to Google's new advertising model. That's all water under the bridge now.
Archive.org is a good history keeper of the Web. As pages and entire sites change over time they keep a log of these variations. For someone in testing, who is constantly in rabbit hole analytics and variables - this can be a great resource to compare different site designs, SEO schemes, page layouts, etc. with a visual history of the Website.
Aim small, miss small.
It's a quote from "The Patriot" with Mel Gibson and referenced in a speech I attended at internet conference way back. But it holds true for successful testing and Web updates in general.
Big and complex legacy tests exist at every company - Everest-like projects that dare corporate employees to ascend. Did you see the movie Everest? My advice is to stay clear of these projects. Nothing good ever came of them. Did you hear the story about that huge project that person took on and everything went well for them? Nope. Guess why?
You might be tempted to be the first ever to prove something out, go "all in" on a project. But, aside from ego, there's no reason. Chop the big projects up into shorter timelines and smaller tests, where you can nimbly adjust to the data you see on a daily basis.
Trigger Vs Typers
With Google Analytics Goals or Events, reporting is usually black and white - sessions either trigger a page or event or not. With people actually reporting the event, reporting is fifty shades of grey - like an MLB umpire calling a strike.
I recommend always having a black and white trigger or data set to compare to the grey areas to have an "all things equal" quantitative analysis over time. However, getting qualitative feedback from the people typing or manually scoring is another way to slice the data at a deeper level - albeit harder to compare year over year.
Annotate
Setting weekly reminders to record the events that took place that week in tools like Google Analytics annotations is a good habit. It's amazing to look back over months of work and see so many updates, but even better to look back at the data and realize why the data shifted during that period.
Make Mistakes
Finally, don't be afraid to try something new and make mistakes. Welcome mistakes in testing. As Albert Einstein said, "Anyone who has never made a mistake has never tried anything new."
